Control of power converters in power transmission networks

The subject matter herein relates generally to the field of power transmission networks and more specifically to the control of power converters in power transmission networks.

In high voltage direct current (HVDC) power transmission networks, alternating current (AC) power is typically converted to direct current (DC) power for transmission via overhead lines, under-sea cables and/or underground cables. This conversion removes the need to compensate for the AC reactive/capacitive load effects imposed by the power transmission medium, i.e. the transmission line or cable, and reduces the cost per kilometre of the lines and/or cables, and thus becomes cost-effective when power needs to be transmitted over a long distance. DC power can also be transmitted directly from offshore wind parks to onshore AC power transmission networks, for instance.

The conversion between DC power and AC power is utilised where it is necessary to interconnect DC and AC networks. In any such power transmission network, power conversion means also known as converters (i.e. power converters in converter stations) are required at each interface between AC and DC power to effect the required conversion from AC to DC or from DC to AC.

The choice of the most suitable HVDC power transmission network or scheme depends on the particular application and scheme features. Examples of power transmission networks include monopole power transmission networks and bipole power transmission networks.

The dynamic exchange of active and reactive power between a power electronic converter interfaced system, such as an HVDC transmission link and the AC power system or grid itself, is primarily governed by a control algorithm employed by the power electronic converter. More specifically, synchronous grid forming (SGFM) control is gaining in interest and popularity owing to its strengthening effect on the stability of the overall power system.

A typical characterization of SGFM control is to make a power electronic converter behave as a three-phase, positive phase sequence, AC voltage source, that resides behind an impedance. The power electronic converter operates at a frequency that is synchronous with other SGFM sources connected to the same power system. The use of SGFM control brings benefits in scenarios where an AC power system perturbation materializes. In response to such a perturbation, the power electronic converter would inherently exchange transient current that would counteract the voltage and/or frequency change experienced at the point of connection of the converter to the AC power system.

However, unlike other SGFM sources such as synchronous generators, power electronic interfaced systems, including power converters in HVDC converter stations, need to obey much stricter current limits. Hence, SGFM converter control methods and apparatus must not allow steady state or transient currents to exceed these limits. In order to achieve this, SGFM control methods and apparatus must act to modify the output voltage profile of the power converter should a risk of non-compliance with the current limits arise.

An example of a scenario whereby the current limits of a power electronic converter could be exceeded is that of an insulation fault occurring in a connected AC power system. This may result in a relatively large voltage deviation at the point of connection of the power electronic converter and the AC power system. The larger the voltage difference between the point of connection and the voltage behind the converter impedance, the higher the current exchange will be.

A three-phase system is designed to operate as a balanced system comprising only positive phase sequence voltage/current. However, when an AC power system undergoes an asymmetric fault, such as a single-phase to earth fault, a large negative phase sequence voltage component may be introduced and experienced at the converter point of connection to the AC power system. In addition, a substantial reduction of the positive phase sequence voltage component may be experienced. If the converter was to maintain purely the positive phase sequence AC voltage profile, as desired, for instance, in SGFM control modes, the negative and positive phase sequence voltage components resulting from the fault would likely drive large negative phase and positive phase sequence current flows within the circuit. Combined with other current components, this could potentially overload at least a portion of the semiconductor devices within the power converter.

More specifically, with regard to modular, multi-level voltage sourced converters (MMC VSC) that are typically employed in HVDC systems, this unmitigated exchange of negative phase and positive phase sequence AC currents, combined with the DC and converter circulating currents, could overload the transistors (insulated-gate bipolar transistors (IGBT), for instance) in the sub-modules of a converter valve or group of valves.

Therefore, there is a need for a method of controlling a power converter that, under any viable AC grid condition, limits the short and long term amplitude profiles of the AC current, to values acceptable for all power components of the power electronic interfaced system. A challenge arises however, in view of prior art approaches, which independently treat the positive and negative phase sequence components. More specifically, the positive phase sequence component, which governs the power flow during normal operation, is often prioritized. However, the electrical current capability of a power converter is based on the total instantaneous value of the AC currents and valve currents. Put differently, it is the instantaneous sum of all of the electrical current components, that can determine whether a power converter is being overloaded. By way of example, considering the AC current only, the highest instantaneous value an AC circuit element would be exposed to on the converter side of a coupling transformer, would predominantly be the peak of the fundamental frequency current waveform that results from the superposition of the positive and negative phase sequence components. Employing a control algorithm that separately regulates the positive and negative phase sequence components, does not consider this superposition effect, and hence some other process must coordinate the regulation such that the highest resulting total AC current amplitude does not exceed the instantaneous value that would overload the power converter and associated power electronics.

In addition, from an AC network (i.e. a power grid) perspective, the coordination of the regulation of positive and negative phase sequence currents will also affect how the power electronic interfaced system would contribute, during disturbances, towards positive phase sequence voltage amplitude enhancement, versus voltage balancing (i.e. restoring voltage symmetry).

A prior art approach to mitigating these issues is to regulate the flow of the negative phase sequence current to zero through the use of a negative phase sequence current feedback controller. A controller of this type measures the power converter AC current and decomposes said current into sequence components in a manner well understood in the art. Then, the controller determines a desired converter output negative phase sequence voltage to drive the measured negative phase sequence current to zero Amperes. In this way, the controller can employ any viable positive phase sequence current limiting strategy without any need to coordinate with the negative phase sequence control algorithm.

The inventors have realized that whilst this prior art approach obviates the need for coordination of positive phase sequence control with negative phase sequence control, it renders a power electronic interfaced system transient AC output current undesirable from a system perspective. More specifically, the resulting transient AC output current would always be positive phase sequence only, regardless of the type of the AC system fault or disturbance that led to the initial unbalance. As a consequence, the power electronic interfaced system (the power converter) being controlled in a such a manner, would not contribute to voltage balancing during asymmetrical faults. Indeed such a control method would likely cause overvoltage on healthy phases of a multiphase supply during fault events. Hence, such an approach could unnecessarily boost the positive phase sequence voltage amplitude enhancement when it is not needed, whilst ignoring the need to restore voltage symmetry. Furthermore, positive sequence only current injection during asymmetrical faults could cause AC protection systems to operate incorrectly.

An alternative prior art approach to mitigating these issues is to switch the power converter into a current control mode for both positive and negative phase sequences when an overcurrent condition is detected. This control method involves sampling current and resolving the current into positive and negative phase sequence components. Then, by superposition, the controller executing the control method determines the largest peak value of the converter output AC current that would result from such symmetrical current components. Finally, based on a difference between this value and a predetermined limit value for the output AC current, the controller determines a reduction factor to apply, equally, to both the positive and negative phase sequence components. These scaled quantities then become the reference for the controller. By driving the output AC current to these references, the largest peak value amongst any one of the phases in a multi-phase system should be equal to the targeted AC peak limit value.

The inventors have realized that whilst this prior art approach tends to guarantee the safe operation of a power electronic converter, whilst providing some negative phase sequence current in addition to positive phase sequence current, a problem arises during asymmetrical AC system faults/disturbances, where the approach may still negatively impact AC system voltage profiles and protection systems. This is because, on first activation of the current limiting control method, the apparent positive and negative phase sequence currents at that instant in time are being used to determine the reference values for future current limiting. The initial step of resolving the measured AC current to symmetrical sequence components (positive and negative phase components, for instance) is burdened with delay and initial calculation errors. Thus, such first impression current references may not be beneficial values from an AC system stability perspective. By way of example, it may follow that the initial current references exacerbate voltage imbalance at the point of connection of the power converter to an AC power grid. This could turn into a positive feedback situation, locking the power converter into a sustained undesirable AC current response through the entire AC system disturbance duration. The risk of such a response is increased for weak AC systems, where the power converter AC output would have the dominant impact on the voltage profile at the point of connection. This is particularly relevant given that greater use of SGFM control is stemming from decreasing voltage stiffness of AC grids (i.e. grids becoming progressively weaker).

Hence, it is desirable to provide a method of controlling a power converter that mitigates these issues.

According to a first aspect of the invention, there is provided a computer-implemented method of controlling a power converter in a power transmission network, the power converter having an alternating current (AC) side electrically connected to an AC network at a point of connection, the method comprising: receiving a first amplitude limit value for a first AC current, the first AC current being output from the AC side of the power converter; receiving a second amplitude limit value for a negative phase sequence component of the first AC current, wherein the second amplitude limit value is less than or equal to the first amplitude limit value; measuring the negative phase sequence component of the first AC current to provide a measured second amplitude; regulating the negative phase sequence component to flow with a regulated second amplitude by: if the measured second amplitude is less than the second amplitude limit value, setting the regulated second amplitude to be equal to the measured second amplitude; if the measured second amplitude is equal to or greater than the second amplitude limit value, setting the regulated second amplitude to be equal to the second amplitude limit value; regulating a positive phase sequence component of the first AC current to flow with an amplitude not exceeding a regulated third amplitude by: setting the regulated third amplitude as a function of the first amplitude limit value and the regulated second amplitude, such that the regulated second amplitude and the regulated third amplitude, when combined, provide a first amplitude for the first AC current that is substantially equal to the first amplitude limit value.

The inventors have found a solution that tends not to limit the flow of positive and negative phase sequence currents in all circumstances, but moreover restricts their respective amplitudes in extreme cases to dynamically calculated limits. More specifically, the regulated second amplitude of the negative phase sequence current component is limited to a fixed, but settable, value, referred to herein as the second amplitude limit value. The regulated third amplitude of the positive phase sequence current component is limited to a dynamically calculated value, referred to herein as the regulated third amplitude, which is a function of: a fixed, but settable, total AC current amplitude limit (herein referred to as the first amplitude limit value); and the regulated second amplitude of the negative phase sequence (which may be the second amplitude limit value or a measured second amplitude of the negative phase sequence component). The function itself determines a regulated third amplitude that when combined with the regulated second amplitude, tends to give the first AC current a first amplitude that is substantially equal to the first amplitude limit value. Put differently, the function finds a positive phase sequence current that together with the negative phase sequence current (having the measured amplitude or the limited amplitude, whichever is smaller), would result in the maximum allowed total AC peak current in at least one electrical phase of an AC system.

Regulating the negative phase sequence in this manner, tends to allow for the negative phase sequence response from the power converter to be independent from the positive phase sequence response. Hence for as long as the measured second amplitude of the negative phase sequence component is below the second amplitude limit value, the power converter will tend to behave as a purely positive phase sequence AC voltage source, allowing the negative phase sequence current to flow naturally and in proportion to the degree of voltage unbalance caused by an AC system disturbance/fault. The second amplitude limit value should be set such that it is less than or equal to the first amplitude limit value. Put differently, the value of the negative phase sequence current limit should be set such that the maximum negative phase sequence output current amplitude does not exceed the maximum total AC peak current limit value.

Regulating the positive phase sequence in this manner, tends to ensure the positive phase sequence response from the power converter is dependent upon the negative phase sequence response from the power converter. Hence, the positive phase sequence current amplitude when combined with the negative phase sequence current amplitude, tends not to yield a total AC current amplitude that exceeds the total current limit for the power converter. Instead positive phase sequence amplitudes are allowed up to an amplitude limit that is defined by the regulated third amplitude.

In a first example useful for understanding the invention, a scenario typical for normal AC grid operation is where no AC voltage unbalance exists at the electrical point of connection of the power converter to the AC network (for instance an AC power grid). In such a scenario, the converter tends to behave as a purely positive phase sequence AC voltage source, and hence would not output any negative phase sequence current to the AC grid. In this case, the regulated third amplitude of the positive phase sequence component will be calculated to allow positive phase sequence current to flow with an amplitude up to the first amplitude limit value. Put differently, the whole AC ampacity of the power electronic interfaced system can be utilized for the positive phase sequence current response, should it be required.

In a further example, useful for understanding the invention, in a scenario where some AC network voltage unbalance is present at the point of connection with the power converter, but the resultant measured second amplitude of the negative phase sequence component is not greater than the second amplitude limit value, the negative phase sequence current will still be permitted to flow naturally. However, the regulated third amplitude of the positive phase sequence component will be calculated such that the positive phase sequence component can utilize only the remaining portion of the first amplitude limit value (the remaining portion of the AC ampacity of the power converter) having taken into account the negative phase sequence current flow.

In a further example, useful for understanding the invention, there is a scenario wherein a negative phase sequence component of the first AC current has a measured second amplitude that is greater than the second amplitude limit value. In this scenario, the regulated second amplitude of the negative phase sequence component will be limited to the second amplitude limit value. The regulated third amplitude hence amounts to the remaining portion of the first amplitude limit value (the AC ampacity of the power converter) after taking into account the second amplitude limit value. Again, the total AC current limit of the power converter is respected.

It may be, for instance, that the second amplitude limit value has been set to correspond to the maximum proportion of the total output current any SGFM source could contribute in any situation, such that the proposed limit coordination method would inherently scale both the positive and negative phase sequence components when operating at the limiting condition. Equal priority tends to thus be given to AC grid voltage balancing and amplitude management. Moreover, for any type of grid disturbance, the regulated/limited converter current would have a similar composition in terms of the positive and negative phase sequence components as from an unrestricted SGFM source, limited in amplitude. Hence, for the example of an AC grid insulation fault, a power electronic interfaced system when controlled in accordance with the invention, would provide the first AC current having a profile more recognizable by conventional AC system protection systems.

It may be, for instance, that the second amplitude limit value has been set (for instance by a user) to correspond to a value less than the maximum proportion of the total output current any SGFM source could contribute in any situation. This would give higher priority to the AC grid voltage amplitude enhancing function of the power converter, vice voltage symmetry restoration. The invention disclosed herein allows greater user configurability in restricting the maximum current contribution of the native phase sequence current so that it is acceptable for stable and safe operation of the power electronic conversion technology.

Some embodiments comprise measuring the negative phase sequence component to provide a measured second phase angle; measuring the positive phase sequence component to provide a measured third amplitude value and measured third phase angle; and wherein the function is further a function of an angular difference between the measured second phase angle and the measured third phase angle.

The phase angle of the respective positive and negative phase sequence components at an instant in time will tend to affect the resultant total amplitude when the components are combined. It may not necessarily be a simple algebraic sum of the regulated second and third regulated amplitudes that is desired to achieve a combination of limit values that provide the first amplitude value equal to the first amplitude limit value (as would be the case if the angular difference was zero). By further considering the angular difference in the function itself, the regulated third amplitude tends to be determined taking account of the trigonometric relationship between the negative phase sequence and positive phase sequence components. This tends to result in a greater usage of the total AC ampacity of the power converter.

The steps relating to measurement or measuring may utilize suitable measurement means. For instance, a voltmeter or ammeter may be utilized to measure voltage and current directly. The measuring steps may therefore comprises receiving the measured values and/or controlling a measurement means to obtain the measured values.

In some embodiments, the regulating the positive phase sequence component to flow with the regulated third amplitude further comprises: projecting the measured negative phase sequence component to at least one second phase vector in a reference frame of a fundamental frequency of the first AC current; and projecting the measured positive phase sequence component to at least one third phase vector in the reference frame.

When considering the electrical current capability (ampacity) of a power converter, it tends to be the case that the instantaneous sum of all current components is of importance. Moreover, for the AC contribution, the highest instantaneous current value experienced by an AC circuit element on the power converter side, tends to be, predominantly, the peak of the fundamental frequency current waveform that results from the superposition of the positive and negative phase sequence components. By projecting these components as phase vectors in a reference frame at the fundamental frequency, the regulated second amplitude and regulated third amplitude can be resolved for the fundamental frequency waveform, and hence, in a manner that considers the dominant contribution to the instantaneous current value experienced by the power converter. Phase vectors, also known as Phasors, are a convenient representation of AC power components having amplitude and phase.

In some embodiments, the first AC current is a multi-phase current; the at least one second phase vector comprises a second phase vector for each electrical phase of the multi-phase current; and the at least one third phase vector comprises a third phase vector for each electrical phase of the multi-phase current.

In some embodiments, the regulating the positive phase sequence component to flow with the regulated third amplitude further comprises: calculating, for each electrical phase of the multi-phase current, a nominal angular difference between the measured second phase angle and measured third phase angle of the corresponding second and third phase vectors, thereby generating a plurality of nominal angular differences; and selecting, as the angular difference, the minimum nominal angular difference from the plurality of nominal angular differences.

The invention as disclosed herein is of particular relevance to multi-phase, for instance three-phase, power converters and power systems. By projecting the sequence components of the plurality of electrical phases of a multi-phase system, into the reference frame, it can be determined which pair of second and third phase vectors have the smallest phase angle difference.

In some embodiments, the function is provided by Equation 1 as follows:

wherein

is the regulated third amplitude, lis the first amplitude limit value,

is the regulated second amplitude, φis the angular difference.

The function provided by Equation 1 can be compared to finding a magnitude of a vector in the direction of the third phase vector (corresponding to the measured positive phase sequence component) that when added to the second phase vector (corresponding to the measured negative phase sequence component), will make the tip of a phase vector corresponding to a total AC current fall exactly on a circular boundary, centered at the origin of the reference frame, and having radius equal to the first amplitude limit value (the total AC current amplitude limit, for the given electrical phase).

In some embodiments, the first amplitude limit value is a predetermined first amplitude limit value, preferably a maximum instantaneous total current limit for the power converter; and/or the second amplitude limit value is a predetermined second amplitude limit value, preferably a maximum negative phase sequence current limit for the power converter.

The first amplitude limit value and the second amplitude limit value being predetermined, but settable, allows the control method disclosed herein to be configured according to a given power transmission scheme. For instance, the values can be set based on knowledge of the limitations/protection/overload thresholds for power electronics in the power transmission network (particularly, the power converter). The values can also be set based on the desire to achieve voltage balance.

In some embodiments, the power converter comprises a voltage sourced converter (VSC), preferably a modular multi-level converter (MMC).

The invention disclosed herein allows greater user configurability in restricting the maximum current contributions of phase sequence currents so that the current contributions are acceptable for stable and safe operation of the power electronic conversion technology. For instance, for MMC VSC HVDC converter stations, a too high level of negative phase sequence current may disrupt valve energy management processes, or could require overdesign of circuits (e.g. by inclusion of additional valve sub-modules).

In some embodiments, the power transmission network is a high voltage direct current (HVDC) power transmission network.

HVDC power transmission networks utilize power electronic converters and control methods such as SGFM for which the problems hereinbefore mentioned occur. The invention described herein provides a solution to these problems.

In some embodiments, the AC network is an AC power grid.

As discussed herein, faults such as insulation faults to earth in an AC power grid can cause voltage unbalance at the point of connection of an AC power grid to a power converter. This can result in large negative and positive sequence current exchange, risking damage to power electronics. The invention discussed herein provides a solution that protects power electronics from protection operation, whilst still allowing negative and positive sequence current to flow in a coordinated manner.

In some embodiments, the method is for use in SGFM.

In SGFM, power electronic converters are configured to behave as voltage sources. However, said converters have intrinsic limitations with respect to current flow that other synchronous devices may not be so susceptible to. The invention described herein considers such limitations and controls the converter in a manner that regulates negative and positive phase sequence current flow to safe levels for converter hardware, when required.

According to a second aspect of the invention, there is provided a controller for controlling a power converter in a power transmission network, the controller comprising: a memory; and at least one processor; wherein the memory comprises computer-readable instructions which when executed by the at least one processor cause the controller to perform the method of the first aspect of the invention.